Nobody said the Nobel Committee was infallible. It did, after all, give Henry Kissinger the peace prize in 1973. But the folks in Stockholm have traditionally been conservative about whom they bestow scientific awards upon. Albert Einstein got his Nobel in physics 16 years after he published his work, but the committee declined to endorse that reckless relativity stuff.
What, then, is one to make of the Nobel Prize in physiology or medicine awarded to neurologist Stanley Prusiner this week? Prusiner's hypothesis is that fatal brain maladies such as mad-cow disease and Creutzfeldt-Jakob disease are caused not by viruses, bacteria, fungi, or some other mundane agent but by something startlingly new that he has discovered--mutant, rampaging proteins known as "prions," short for "proteinaceous infectious particles."
But do prions cause these diseases? In the past year, Science and Nature, the two most distinguished scientific journals, have published three major papers suggesting that the causative agents of these diseases are not prions, and that Prusiner's 15 years of prion research is simply wrong. The latest of these papers was published last week, in Nature's Oct. 2 issue.
Before moving on, a quick course in molecular biology: All living creatures, from viruses on up, pull off the feat of self-replication by encoding the necessary information in nucleic acids--in particular the double-stranded DNA discovered by Crick and Watson or, for a few renegade viruses, single-stranded RNA or even single-stranded DNA. These nucleic acids then code for and generate proteins, which are the stuff we're made of. Prusiner's proposition is that it is proteins, not "slow viruses," that are the infectious agent in mad-cow-like diseases. This is what makes Prusiner's hypothesis so radical: Prions would be the only proteins on the planet that reproduce--not to mention infect and kill animals--and thus the only known exceptions to the rules of the Crick-Watson paradigm of molecular biology.
Prusiner's proposition has been controversial from the get-go. The researcher who did Prusiner's lab work at the University of California at San Francisco quit over the publication of Prusiner's very first prion paper in 1982, arguing that Prusiner was overinterpreting the available data to push the prion hypothesis.
Over the next 15 years, Prusiner won over virtually everyone to his prion hypothesis--the lay press, the scientific press--but not the researchers in his field. He got his share of bad press, for which I take--and am given--entirely too much credit (read my December 1986 feature story in Discover by clicking). By 1985, when Prusiner's own papers were still suggesting that the prion hypothesis was at best a long shot, he won a $4-million congressional award "to determine the structure of prions and how they cause disease." In 1991, Prusiner reported at a major conference that he had proved that the infectious agents of these diseases were proteins free of nucleic acids. In particular, he had created mice with a genetic mutation that caused what was a normal protein--the prion protein, in Prusiner's lingo--to become abnormal and produce disease. He then took brain matter from these mice and injected it into new mice, which promptly got sick, showing that no viral particles were necessary to transmit the disease. But there was still no paper proving the results to the scientific community.
T wo years later, when he presented the same mice work at another conference, the news pages of Science and Nature wrote it up as if it cinched the prion hypothesis. Although Prusiner's work had not been replicated by anyone and he had still not published these supposedly seminal findings that prions cause disease in a peer-reviewed journal, he won the prestigious Albert Lasker Award in October 1994. (The Lasker Award is considered a short-list for the Nobel.) When he finally published his proof-of-prion paper, it was only after it had been rejected by the journal Cell. Prusiner managed to find a home for it in the Proceedings of the National Academy of Science, where it wouldn't have to be peer-reviewed. The paper was largely ignored, and it was dissed even by Prusiner, who admitted to a New York Times reporter that it was uncompelling. Scientists who read the paper suggested that its findings could be explained by contamination, which is to say, by sloppy laboratory procedures. But four months later, Prusiner was still describing the mice work to Scientific American's lay readers as "a persuasive experiment."
Prusiner's boilerplate response to prion critics has been that if mad-cow disease or any of the other "prion diseases" is caused by a virus, then surely that virus would have been discovered by now. The fact is, it's damned hard to find a virus in a mishmash of animal brains, which is where you have to look. One reason a virus hasn't been found--if indeed a virus causes these diseases--is that no one is doing the laborious and expensive work to find it. It can take researchers decades to find culpable viruses--hepatitis C is a famous example. But at least those researchers got funded to look, which has not been the case in the prion field. Prusiner has received in the neighborhood of $40 million in funding from the National Institutes of Health since 1985. Qualified critics who request money from the NIH to look for viruses are told by NIH bureaucrats that if the virus turns out not to exist, then their study will have been "of insufficient significance and scientific merit" and thus not worth doing.
Despite the vindication offered by Stockholm, the prion hypothesis is still rife with loopholes. For instance, the diseases that allegedly are caused by prions come in a few dozen different strains, the same way that dogs come in different breeds. It's easy to imagine variations in viruses or bacteria, because they contain nucleic acids, which encode for variations. But even Prusiner hasn't been able to explain how a protein that has no nucleic acid could encode for the variations. His own grant proposals, available on the Web through the NIH CRISP database, are evidence of the problem. One describes the problem of prion strains as a "fascinating conundrum," while another explains that the goal of the research project is to find out "whether the strains differ in the properties of the scrapie form of the prion-protein ... or [whether] a second component is responsible for strain specificity, the obvious candidate being a nucleic acid."
Strains aren't the only problem with the hypothesis. Prusiner has yet to show, for instance, that a protein sans nucleic acid can be infectious, and consequently, he has invoked the potential involvement of yet another agent in the disease process (although he insists it has no nucleic acid and calls it "Protein X"). Some of his fellow prion researchers suggest it might be a "viral co-factor," which is doublespeak for saying that the prion ain't the infectious agent, a virus is.
One member of the Nobel Committee says his colleagues were aware of unanswered questions in the prion hypothesis but awarded Prusiner the prize in recognition of the wealth of information he has unearthed on mad-cow-like diseases. But if it turns out that viruses do cause the diseases, then Prusiner will have won the prize for the discovery of something spectacularly wrong.
Good science, not just Nobel Prize-caliber science, depends on hypothesis and test, and then the rigorous demonstration that the preferred interpretation of the data was the only interpretation. In other words, remarkable results demand remarkable evidence. In the case of Prusiner's prize, the Nobel Committee has settled for enthusiasm and single-mindedness.